通过免疫组织化学对腺体细分的自动化基础真理注释.
Tushar Kataria1, Saradha Rajamani1, Abdul Bari Ayubi2
1Kahlert School of Computing, University of Utah, Salt Lake City, Utah; Kahlert School of Computing, Scientific Computing and Imaging Institute, University of Utah, Salt Lake City, Utah.
概括
使用免疫组织化学 (IHC) 标签对结肠病理的自动腺注释显著减少了手工劳动. 这种方法可以为诊断炎症性肠病和癌症提供准确的深度学习模型训练.
科学领域:
- 计算病理学计算病理学
- 数字组织病理学 数字组织病理学
- 机器学习在医学中的应用
背景情况:
- 对结肠腺的显微镜评估对于诊断炎症性肠病和癌症至关重要.
- 深度学习模型提供了对腺组织架构的系统,可重复和定量评估.
- 用于深度学习的组织病理学幻灯片的手动注释是耗时和昂贵的.
研究的目的:
- 开发一种自动化的方法,用于在H&E染色的幻灯片中生成结肠腺的基本真相注释.
- 使用免疫组织化学 (IHC) 标签将腺体面膜转移到H&E图像中,以进行深度学习模型训练.
- 与手动注释相比,评估自动注释的性能,并提高模型的概括性.
主要方法:
- 开发了一个图像处理管道,将腺体面罩从KRT8/18,CDX2或EPCAM IHC转移到共同注册的H&E图像中.
- 使用自动化IHC衍生注释训练深度学习模型.
- 将模型性能与内部和公共数据集的手动注释进行比较.
- 提出了一种数据采样技术,以适应模型的新数据源.
主要成果:
- EPCAM IHC提供了腺体轮,与手册注释密切匹配 (子=0.89),并且对炎症有弹性.
- 用10%的注释病例进行训练的模型在公开数据集上取得了高绩效 (迪斯分数为0.902和0.89).
- 使用特定细胞类型IHC标记器的自动注释显示了与手动注释相比的性能.
结论:
- 使用IHC标签的自动腺注释可以有效地取代数字病理学中的手动注释.
- 拟议的方法有助于开发强大的深度学习模型,用于结肠癌和炎症性肠病诊断.
- 一种简单的数据采样技术可以提高模型在各种数据集中的适应性,从而提高概括性.
更多相关视频
05:45Quantitative Immunohistochemistry of the Cellular Microenvironment in Patient Glioblastoma Resections
Published on: July 31, 2017
9.7K
13:01Industrialized, Artificial Intelligence-guided Laser Microdissection for Microscaled Proteomic Analysis of the Tumor Microenvironment
Published on: June 3, 2022
3.8K
相关概念视频
Immunogold Electron Microscopy
4.1K
Immunoelectron microscopy utilizes immunogold labeling of endogenous proteins with specific antibodies to detect and localize these proteins in cells and tissues. The procedure provides insights into the distribution and quantification of protein under different stimulation conditions offering clues about their functions. Conjugating highly electron-dense gold particles with primary or secondary antibodies allow antigen detection on and within cells, with high resolution and specificity.
4.1K
Immunocytochemistry and Immunohistochemistry
11.4K
Immunocytochemistry (ICC) and immunohistochemistry (IHC) are techniques that use antibodies to check for specific proteins or antigens in a sample. The technique was first published by Albert Coons in 1941 to detect the presence of pneumococcal antigen in tissue sections from mice infected with Pneumococcus. Immunocytochemistry helps localization of proteins or antigens in individual cells like blood cells, stem cells, etc., while immunohistochemistry does the same for tissue samples.
These...
These...
11.4K
